DESCRIPTION The investigators propose to use data from the Epidemiologic Follow-up Study of the National Health and Nutrition Examination Survey (NHANES) to develop models to predict morbidity and mortality as a function of risk factors for CAD. Bivariate dependent survival distributions will be examined.